Natalia Shmatko

RESEARCH TEAMS’

HUMAN CAPITAL

BASIC RESEARCH PROGRAM

WORKING PAPERS

SERIES: SCIENCE, TECHNOLOGY AND INNOVATION

WP BRP 32/STI/2014

This Working Paper is an output of a research project implemented

at the National Research University Higher School of Economics (HSE). Any opinions or claims contained

in this Working Paper do not necessarily reflect the views of HSE.

Natalia Shmatko1

RESEARCH TEAMS’ HUMAN CAPITAL

Adopting a Bourdieusian perspective, human capital of research team in the paper is

understood as the configuration of the active properties of individual team’s members and the

distribution of differences of their active properties. The paper describes a research team as an

ensemble of social distinctions determined by a distribution of “active properties” of the team

members (educational, research, administrative and media characteristics) and analyses their

empirical distribution.

The study is based on the consideration of sociological factors of research teams’

performance. The paper discusses in depth the performance of research team (from a sociological

point of view) and the relationship between the performance and the social management efficiency.

It is prouved that the performance depends on the efficiency of human capital management research

team. These distributions show how holistic properties of the field develop, how they relate to

individual agents’ properties and how individual researchers “fit” into science.

JEL: I2, I28, J24, O32

Keywords: human capital, scientific capital, research and development, laboratory,

performance, efficiancy, research management

1 National Research University Higher School of Economics, Institute for Statistical Studies and

Economics of Knowledge, Department for Human Capital Studies, Head of Department, E-mail:

nshmatko@hse.ru

3

Introduction

Lately, economists, politicians and sociologists have been paying increasing attention to the

activity of research groups and individual researchers. This is undoubtedly related to modernization

and development of innovation economics. Although such activities have been studied thoroughly

yet since the 1960s, so far neither conventional research methods nor standard analysis procedures

have been established. Most progress has been attained in the analysis of economic aspects,

resources and results of the research groups’ activity. Among its sub-topics, effectiveness is the one

that attracts most attention[Gokhberg and all, 2013]. It is usually defined as “a measure of the

aggregate of all results achieved during a research or a measure of goal achievement and outcome of

an R&D project.” Research efficiency is also studied rather often and is referred to as “a measure of

correspondence between effectiveness of R&D activities and consumed resources”[OECD, 2002].

Human capital is another important aspect often considered to be an integral part of the

analysis of organizational activities, including research teams’ activity. It is usually believed to be

equal to a number of economic indicators of efficiency, but can also be regarded as a form of

investment in staff education and professional training.

Professional training and the timely identification of the demands of new skills assume

increasing importance as the technological environment is constantly renewed. This renewal is a

result of technological, product-related and other innovations or their applications, all of which lead

to the emergence of new requirements for personnel knowledge and skills. Unstable markets,

globalisation and short product life cycles force companies and employees to master new skills to

maintain a competitive advantage [Gackstatter et al, 2014]. Since the 1980s, Lundvall has stressed

the close connection between innovations and human capital of institutions among with new forms

of training needed for their implementation, namely, interactive learning [Lundvall, 1985, 1992,

2002].

Policymakers should develop special programmes to improve the ability to innovate

[Meissner et al 2014]. The development of this ability in researchers, engineers and technicians

should be part of larger programs for the adoption of new technologies and products. The ability to

innovate is determined by the qualifications, skills and mobility of staff. In turn, the mobility of

researchers and the transfer of their skills and competencies play a crucial role in the diffusion of

innovation.

4

We considere the performance of research team (from a sociological point of view) as

directly related to social management efficiency. Performance depends on the efficiency of human

capital management research team.

This study is based on the analysis of sociological factors of performance of research teams.

Economic indicators of performance weren’t considered in this project.

Methodology and Approach

Human capital is defined as “an aggregate of knowledge, skills, motivations and energy of

an individual that can be used for production of goods and/or services.” [Romer, 1990] This

definition focuses on future profits gained through investments in human resources at all levels.

Such understanding of human capital is linked to the concept of the scientific and technical

revolution.

However, even if we adopt a broader concept of human capital, such approach to the analysis

of the activity of research groups does not seem to us fully relevant as it disregards several

significant aspects of organization and functioning of science, a scientific research and researchers.

We are convinced that the approach suggested by the genetic structuralism[Bourdieu, 1975,

1988; Karady, 1977, 1979;Lebaron, 1997]is by far more productive. It offers necessary prerequisites

for a complex analysis of science and its institutes both on macro- and micro-level. Individual

researchers are considered as individual agents (actors) producing scientific knowledge, and R&D

laboratories and institutes represent collective agents. According to the postulates of the genetic

structuralism, a basic structure that is a condition and a prerequisite for scientific practices is

referred to as the scientific field[Bourdieu, 1984; Barnes, 1982; Bloor, 1976, 1983].It is a

multidimensional social space where each individual or collective agent (such as research

laboratories) occupies a certain position and is involved in competition for scientific recognition.

The scientific field can also be regarded as a distributed system or a social network

[Chalmers, 1990; Collins, Pinch, 1982] that consists of positions held by individual and collective

agents who interact between themselves, and evolution of each agent is determined by his

neighbours’ behaviour.

Agents (especially collective agents, such as, for example, research institutes or editorial

boards of scientific magazines) are characterized by complex internal structure. However, this

5

complexity does not reveal itself in their interaction and thus in the analysis of the scientific field

they can be considered as relatively simple objects boasting of a small number of effective degrees

of freedom.

The scientific field can be distinguished as an independent subject of study, and we can

name three important reasons for why such a distinction is feasible and productive for the purposes

of our research. First of all, this approach implies relative autonomy of scientific practices from any

external influence, be it political, administrative, state or economic one. Second, there exist stable

invariants of scientific practices that can manifest themselves as norms, standards, traditions and

movements. Third, all resources needed for production of a scientific discourse as a specific kind of

a social game are limited: there is always a moderate number of R&D institutes, positions, awards,

magazines, grants and so on. Scanty possibilities lead to inevitable reduction in potential complexity

of researchers’ practices to a restricted set of scientific positions.

The structure of the scientific field is defined by the distribution of the specific capital

possessed by the agents in the field. Combination of various capitals determines how integral the

field in question is and describes concordant changes of the key features (‘active properties’) of its

agents. Different kinds of capital – cultural, educational, scientific, administrative and social – are

thus regarded as variables that determine the scientific field and all other variables. Distribution of

those kinds of capital (that is, relations between them) subordinates the characteristics that can be

measured directly. As a consequence, we can substitute a description of a large number of

‘microscopic’ variables by a description of two or three substantial ‘macroscopic’ ones. The

development of the structure of the scientific field is a result of a fluctuation rise in active properties

of researchers, enforced by influences and powers, external to the field itself.

Capital is an analytic category constructed by a sociologist on basis of the empiric data he

possesses with regards to agents’ active properties. Different kinds of capital are to be defined

separately in each case, since they represent invariant field properties, and the latter in turn depend

on authority and power distribution inside the field.

Every field is characterized by a specific capital that determines the functioning of that field

and explains behaviour of individual or collective agents that act in it. For example, as far as the

political field is concerned, its main capital is the political capital, and respectively, the economic or

financial capital dominates over the economic field.

Empirically, each capital is a factor constructed through statistical procedures of

multidimensional scaling [Bourdieu, 2004] that distinguishes between agents in the sample by a

6

certain characteristic. Such a factor can notbe attributed before a statistical analysis has been held;

on the contrary, the former is an independent resut of the latter.

We have to underline two peculiarities of the notion of capital. First of all, since capital is

understood as a configuration of active properties, it therefore implies rejection of one-dimensional

and “linear thinking, which only recognizes the simple ordinal structures of direct determination”

[Bourdieu, 1979, p.119], and urges to analyze interrelated relationships objectified in the field in

their totality. Capital is a system of active properties where each of the properties enforces the rest

of them and this interaction results in total, ultimate interdependence of those properties rather than

the absence of any dependence. Thus, if a sociologist chooses to use capital – a multidimensional

configuration of agents’ properties – as an explanatory principle for scientific practices, instead of

an ordinary understanding of dependence between the agents’ properties, he/she can construct

several aggregate factors that would explain the whole varierty of practices.

Secondly, the notion of capital is closely associated with the concept of field. That implies

that agents’ properties are not all simultaneously operative. Only those properties specifically related

to the field in which they are objectified or internalized: different fields have their own systems of

active properties that equal to capitals. “The specific logic of the field determines those which are

valid in this market, which are pertinent and active in the game in question”[Bourdieu, 1979, p.127].

Capital determines agents’ practices. At the same type, as it is an abstract ‘ideal type’, a

capital never reveals itself in ‘pure state’ per se, as it can only exist in a specific form. Agents’

practices can be explained through capital, but a sociologist cannot study it directly since it has to be

constructed as an analytical abstraction using methods of mathematical statistics.

Our interpretation of the notion of capital largely disagrees with its widely accepted

interpretation offered by T. Schultz or G. Becker and his successors (M. Blaug, R. Laird, J. Mincer).

They believed that the economic factor was the most important one and, more specifically,

attributed a large weight to the profitability of investment in education and professional training for

economically active population [Schultz, 1971; Becker, 1964; Blaug, 1976; Mincer, 1974,

1994].There is no doubt that human capital is not limited with knowledge and human capacity to

work, neither can it be considered just a spending factor; it is a major social condition for

development of science and innovation. We consciously leave aside the investment component,

albeit acknowledging that certain aspects of human capital (active properties of individual and

collective agents) can be enforced through purposeful investment. The latter do not necessarily have

to be centralized or financial.

7

As applied to scientific production, ‘human capital’ is an aggregate of active properties, seen

as resources of scientific practices that regularly return interest to a researcher. That profit can be

denominated in specific stakes of a game in the scientific field in a long run. We distinguish

between individual human capital of an individual researcher and collective human capital of a

research group. In our concept human capital is a characteristic function, which is a function of state

of respective independent parameters (agents’ active properties). Once parameters values are known,

this function and its derivatives describe all social properties of an agent that we are interested in.

Human capital can also be used in order to explain scientific patterns –any sequences of events of

scientific production–, as the distribution of human capital conveys explicitly the state and dynamics

of the production’s social relationship.

Individual human capital of a researcher can be measured by the following groups of

indicators:

Educational properties: basic education, higher and postgraduate education, a scientific

degree, further professional training, etc.;

Scientific properties: affiliation with a scientific movement, a number of publication,

publications translated into other languages, participation in scientific life (conferences,

seminars, etc.), scientific awards and other tokens of recognition from the colleagues;

Administrative properties: positions held in scientific, educational and state institutes;

membership in editorial, academic and scientific boards; participation in expert

examinations, work groups and and commissions, etc.;

Media properties: media publications of an agent or about him/her, including reviews,

public lectures, participation in TV- and radio shows, personal blogs and web-sites.

Respectively, collective human capital is a distribution function of active properties of a

group.

That is why, when speaking about a subject of social administration, we will always be

referring to the human capital of a research group as an invariant that determines conservation of

main properties of a group while the latter undergoes various changes.

Analysis of relationships exisiting within scientific production allowed us to construct R&D

team as a subject of sociologic study. Addressing those relationships again, we decided to

distinguish the most significant ones and established the study procedure. It allowed us to outline

8

the direction of our research and its goals. Then we realized that practical implementation of the

results of our study enables us to map out concrete measures that could contribute to resolving the

problem of rational transformation of research and development groups.

According to Merton [Merton, 1979], scientific communities arise before democratic social

structures. In scientific communities, scholars are united not by mechanical constraints but by a

specific set of values and norms, joint goals, and culture. However, the clich/e of the scientific

community as a microcosm in which gifts of information are exchanged in return for recognition

among equal colleagues [Hagstorm, 1975] has been shown to be a fiction. In fact, in the “pure”

microcosm of even the “purest” science, there always remains an element of intense competition “in

which the specific issue at stake is the monopoly of scientific authority, defined inseparably as

technical capacity and social power” [Bourdieu, 1975, p. 19].

From a sociological viewpoint, the production of modern scientific knowledge occurs in “a social

field of forces, struggles, and relationships that is defined at every moment by the relations of power

among the protagonists” [Bourdieu, 1991, p.3]. The structure of this type of scientific field or, what

amounts to the same, space of relationships is characterized grosso modo by the distribution of

scientific capital between the agents and institutions that operate in the field [Bourdieu, 1997, p. 14–

21]. Roughly speaking, the concept of scientific capital highlights the process of accumulating

specific resources and benefiting from the academic and administrative status that result in both

reputational and power effects. Theoretically, social regularities of the knowledge generation are

produced and reproduced through the distribution of scientific capital and the interests of the

individual and collective agents in this field (cf. [Bourdieu, 1985, p.724-725]). Scientific capital is

an invariable property in the scientific field that is connected to the allocation of specific scientific

power and recognition. Scientific capital takes its form and content from the scientific field within

which it is used. Scientific capital is country-specific and its currency varies across different national

social spaces. According to Bourdieu, scientific capital is a configuration of active properties (active

in the sense that the properties represent a field of forces) that provide the agent with authority,

recognition, influence, and power in a given scientific field [Bourdieu, 2004, p. 55–58]. Scientific

capital tends to set ensembles of active properties of individuals within the social structure of

scientific field.

Bourdieu’s approach to the scientific capital has been empirically tested [Bourdieu, 1988;

Brosnan, 2011; Gartorth, 2011; Hong, 2008; Lebaron, 2001; McGuire, 2011; Panofsky, 2011;

Rouget, 2002].

9

For a critical analysis of this approach, see [Bellotti, 2011; Brubaker, 2005; Camic, 2011;

Coradini, 2010; Grosetti, 1986; Jain, 2013; Sismondo, 2011].In the literature, scientific capital is

also presented [Corolleur, 2004; Bozeman, 2001, 2004; Dietz, 2005; Lin, 2006] as the sum of

knowledge and work-relevant skills, social links, and resources. However, Bourdieu’s version of the

scientific field — which has an integral character and strives to eliminate the contradictions between

micro- and macro-sociological analysis, agents, and structures — seems preferable. Still, we lack a

mathematical model for scientific capital, because the quantification of scientific capital does not

fall outside the parameters of multivariate statistics [Lebaron, 2009, pp. 11–29]. In order to achieve

the better understanding of the notion of scientific capital, in this paper a mathematical modeling

approach is put forward, which describes firstly structural relations between scientific capital and

both the amount and quality of social resources, and secondly probability distributions of scientific

capital.

Bourdieu’s concept of scientific capital exhibits three principal characteristics:

1. Scientific capital expresses the emergent quality of the set of an agent’s active properties. The

agent’s scientific capital is examined as an attribute of unified scientific field. In this sense,

scientific capital is sui generis the characteristic state function, which means that the condition

of the scientific field can be presented as a function of scientific capital.

2. Understanding scientific capital as an integral configuration of active properties is tantamount to

rejecting single-variant analysis based on “linear thinking, which only recognizes the simple

ordinal structures of direct determination, and endeavors to reconstruct the networks of

interrelated relationships which are present in each of the factors” [Bourdieu, 1984, p. 107].

Scientific capital is a system of active properties in which each quality strengthens the others.

3. The active properties are the efficient characteristics “that are selected as principles of

construction” of the scientific field; “are the different kinds” of scientific capital [Bourdieu,

1985, p. 724]. Each active quality receives its value and efficacy from the regularities of the

scientific field [Bourdieu, 1984, p.113].

Active properties are allowed to operationalize as socially significant resources in the

production of scientific knowledge. Here, we refer to resources that regularly result in a specific

gain, a stake in the social game bounded by the scientific field, and that endure for a long period. In

this type of interpretation, scientific capital determines the chances of an agent’s attaining

10

recognition or an administrative post. From this perspective, maximizing scientific capital can serve

as the conceptual focus of the scientific field. (see [Bordley, 1983; Chernyak, 1992]).

Bourdieu’s conceptualization of capital is related to “the set of actually usable resources and

powers” [Bourdieu, 1984, p. 114]. In general, scientific capital may be defined as “accumulated

labor… which, when appropriated on a private, i.e. exclusive, basis by agents or groups of agents,

enables them to appropriate social energy in the form of reified or living labor” [Bourdieu, 2002,

p.280]. Bourdieu’s usage of the concept of “social energy” is a metaphorical but fruitful extension of

scientific capital. According to the general methodological approach, the state of a scientific field is

defined if values to all the quantities used in the sociological investigation are determined.

We proceed from the assumption that the agent’s behavior within the scientific field need not

be rational but that his or her sociological explanation will be rational, which is precisely the

principle of maximizing scientific capital. The agent’s strategy selects high-yield combinations of

the values of active properties. Each strategy seeks to reach a balance of efficiency and the stability

of the agent’s social trajectory within the scientific field: that is, to obtain the best results in different

undefined social situations.

One of our tasks was to adapt a collective agent to the needs of scientific production, and

that brought us closer to the necessity of introduction of social administration defined as directed

purposeful impact on relationships within a collective agent, on its relationships with scientific

production and on the modification of individual agents’ properties.

However, a well-formulated problem of social administration requires concrete information

about its goal, and thus, about new desired properties of a R&D team. Moreover, we shall not forget

that the aim is also determined by the conditions of functioning of a group and external powers that

might have certain impact on the subject of social administration.

In order to take into account the above mentioned conditions and external influences we

have chosen an invariant as the object of social administration, and namely, human capital. It is clear

that human capital in a R&D group is managed through changes in its configuration, that is, through

transformation of the relationships between active properties of individual agents, members of the

group (see [Shmatko, 2011]).

11

Research methodology and sample layout

The task of human capital measurement has been set within the framework of the project

called Factors Of Efficiency In R&D Teams’ Activities.2 During this project, we have studied

research and development groups from the Russian Academy of Sciences (RAN), the Russian

Acdemy of Medical Sciences (RAMN) and the biggest Russian universities, such as the Lomonosov

Moscow State University (MGU) and the Mendeleev University of Chemical Technology of Russia

(MHTI).

We have conducted a survey in 13 R&D groups chosen via purposive sampling. In total, 233

individual researchers were interrogated. The R&D teams were chosen for participation in the

survey according to the following criteria:

Experimental(rather than purely theoretical) research;

Researcharea: chemistry, biochemistry, biology, biomedicine, bioinformatics;

Group size: 10 to 25 researchers.

In such a way we have ensured a relatively homogeneous sample that allowed us to compare

various indicators of each group’s activities and to reveal factors responsible for efficient research

group management.

Apart from this, the study included an experiment that aimed to compare formal and

informal structure in a R&D team. We chose one laboratory where we interviewed both staff

members and out-of-staff employees that regularly worked for laboratory’s projects and/or carried

out various tasks for the group. Such approach allowed us to study an interesting case and compare

structures and efficiency of two forms of organization of scientific research.

Our project did not pretend to evaluate content, efficiency or scientific value of the research

activities of the groups, since it requires a separate study and extensive expert work.

Unlike other empirical sociological studies on science and scientific community that had

been conducted in Russia earlier, our study was based on artefact analysis as those were measurable

indicators: number of diplomas, publications, conference reports, grants, contracts, membership in

expert commissions, etc. rather than scientists’ opinions and self-estimation. It allowed us to

objectify the indicators responsible for measurement and administration of the human capital. The

2HSE Basic Research Programme 2009. This purpose was developed in the frame of following Monitoring survey on

Highly Qualified R&D Personnel (2010-1014).

12

tools for the survey included a comprehensive whole of indicators of practices and properties of bot

a group in its integrity and separate individuals.

Data describing group activities were mainly recorded according to the standard indicators of

scientific activities’ efficiency. For example, the list of “Indicators of Efficiency of Academic

Activities” of the Russian Academy of Sciences includes: reviewed publications in journals,

monographs, patents, conference reports, elaboration of teaching courses, presentation of

dissertations, etc. In our survey, we supplemented the list due to the fact our sample included R&D

groups that dealt with experimental, empirical research and it was important to make a record of

whether they have access to all necessary equipment and facilities required for their experiments.

Therefore, we included in the interview questions on free access for researchers and post-graduate

student employed in various projects to the following:

Engineering, design, R&D and technologic departments and organisations;

Experimental base (R&D production);

Auxuliary departments (workshops, repair and maintenance services);

Scientific equipment and experimental machinery multi-access centre;

Centre for technology transfer;

Centre for innovation and technology;

Industrial park;

Business incubator.

We made a record of educational, scientific, administrative and communicative

characteristics of a head and other employees of every R&D team. All the information gathered later

underwent statistical data treatment and thus we coud determine stable invariants of practices and

properties and display fluctuations typical for groups that work efficiently. The statistical data

treatment procedure included multidimensional scaling of empirical data collected during the survey

and mathematical modeling of R&D teams’ effectiveness.

Results and Discussions

Profiles of R&D teams in the survey

While studying R&D teams, it is important to ask whether a group examined is truly a team,

whether the list of employees corresponds to the list of people actually working on projects of the

team and whether research workers share the same conceptual approach and hold the same views.

13

Our survey revealed that only in three out of thirteen groups the list of employees is a

meaningless paper and the real team consists of research workers from different labs and institutes.

For the majority of the groups the list of employees corresponds to reality and researchers from

other departments might only be invited for short-term collaboration (usually working on a small

specific task) from time to time.

Only in two teams, namely in the group number 3 and the group number 7, their heads

believed that their team does not really consist of like-minded people who share common conceptual

approaches and ideas and stated that their researchers enjoy a large degree of autonomy.

It has to be underlined that R&D teams included in the survey are all different as far as their

social structure is concerned:the list of personnel (its length and positions), age of group leaders,

their experience in team management, publication activity, participation in expert commissions and

other examining nodies, grant activities, etc. (See figures 1-3).

Further analysis proved that presence/absence of post-graduate students and employees

under 35 is an important characteristic that has a big impact on the efficiency of R&D team’s

activities (Figure1). Among the teams in question, only in half of the groups there were more than

30% (which is the sample mean value) of young employees. Respective percentage in the group

number 10 is lower than 20%. It shall be added that younger employees work mainly in research

departments that work on scientific disciplines in conjunction with other scientific disciplines, such

as biochemistry, biomedicine, bioinformatics. The same groups attract the highest number of post-

graduatestudents (Figure2) that combine work on their dissertations with research planned and

required by the laboratory.

Figure 1:Young (under 35) employees’ share in the R&D teams

14

Figure 2: Post-graduate students’ share in the R&D teams

The “number of post-graduate and doctoral dissertations, prepared and presented in a

scientific department” is one of the major ‘traditional indicators’ of effectiveness of R&D teams.

According to this criterion, the most effective R&D teams are the number eleven, the number two

and the number one. Three of the teams (number five, nine and ten) presented two doctoral

dissertations each in the past three years. Two of the thirteen groups included in the survey did not

present any dissertations between 2006 and 2008.

Figure 3: Number of post-graduate and doctoral dissertations presented in respective R&D teams

15

As far as the patent activity is concerned, we have to mention that it is very low in virtually

all R&D teams in question. The only group that shows high activity in this respect is the team

number nine: 25 patents obtained within the last three years. Five of the groups included did not

register any patents in the same period, and four of them registered in the mean time from one to

three patents.

The interviews with the group leaders revealed that they share rather a skeptical attitude

towards patents. Some of them consider it a waist of money (for the

registration/maintenance/renewal of patents); others are convinced it is a meaningless activity since,

from their point view, registration of a patent does not require any significant innovations or

discoveries.

Analysis of data representing publication activity of the R&D teams between 2006 and 2008

(see Figue4) points not only to the fact that the groups have very different levels of this activity but

also indicates that the publication strategy of the teams also differ significantly.

Figure 4: Publication activity structure of the R&D teams, 2006-2008

For example, the team number 12 has not published anything in the interim period. The team

number 4 has presented one article, in collaboration with foreign partners. The team number eleven

features the biggest number of publications, and a vast majority of those are articles published in

16

foreign magazines and publications prepared in collaboration with foreign colleagues. The ratio

between articles published by this team in Russia and abroad is 1:6. A similar trend to seek/prefer

international publications exists in almost a half of the R&D teams: number two, three, eight, nine,

ten, eleven. All of them work in biochemistry, biology or bioinformatics. One “Russian” publication

in these laboratories corresponds to two or three published somewhere abroad.

Other important indicator of publication activity is the ratio between the total number of

publications of the team and the number of publications in the past three years. Our analysis proves

that the groups differsignifictnly in this respect, the most active ones being number 10 and number

11.

Figure 5: Publication activity of the R&D teams

The data concerning publication activity of R&D team members was supplemented with the

citation index provided by the ISI Web of Knowledge, namely, the Science Citation Index Expanded

(see figure 6). According to it, group leaders of the teams number 11, 9 and 13 show the highest

generalized Hirsch h-index: h=31, h=22 and h=22 respectively.

17

Figure 6: Citation indexes of the group leaders of the R&D teams

(ISI Web of Knowledge, 1955—2009)

Grant, or contract, activities of a R&D team are also measured as an indicator of its

effectiveness. It includes a total number of grants obtained from various sources, and contracts for

R&D, engineering, design and technologic projects. We countred separately a number of contracts

for implementation of large-scale projects worth more than 10 mln rubles. The data we gathered

shows that only three groups worked on similar kind of projects in the past three years. Other R&D

teams had smaller grants and contracts, and the teams number four and twelve did not receive any

grants or sign any contracts in 2006-2008, as it was their leaders’ deliberate and conscious choice.

The biggest number of grants and contracts in the interim period was received by the teams

number six and two. For the number six, the main souce of funding came from various contracts

with the industry, while the team number two received their grants from the Russian Foundation for

Basic Research (RFFI).

18

Figure 7: Participation of the R&D teams in the project system

The total number of grants and large contracts received/signed for elaboration and implementation

of the research and development, engineering, scientific design and technological projects

Half of the scientific groups included in our survey enjoy the infrastructure they need: either

directly in their departments or they have access to all the necessary facilities. 10 groups out of 13

use shared scientific equipment multi-access centres. A 70% majority of the teams in question have

an engineering base for experiments or similar auxulisary departments. Access to technology

transfer centres, industrial parks and business incubator were the least demanded facilities.

Space of studied R&D teams

The information we gathered initially was subject to thorough statistical data treatment, in

order to see the preliminary structure of the object of analysis. A factor analysis revealed a certain

structural community of properties of some of the the R&D teams (see figure8).

The space is structured along two main and two auxiliary axes. Aggregate capital is located

along the two main axes, F1 (aggregate scientific capital) and F2 (aggregate administrative capital).

The auxiliary axes structure the space with regards to the expert capital and a new so-to-say-integral

factor which indicates whether a R&D team is geared for integration into the international scientific

society (respectively, a team can be oriented towards achievements of Russian or international

science).

19

Figure 8: Space of the R&D teams

F1 – aggregate scientific capital

F2 – aggregate administrative capital

The R&D teams are located along the axis F1 (which for convenience we called “scientific

capital”) based on several indicators: their publication activity, participation in conferences,

presence and number of employees with scientific degrees and awards, the research profile

(fundamental/applied research). The teams that are located in the left part of the graph, have lower

publication activity, less researchers with scientific degrees and are more interested in elaboration of

applied and practical solutions in comparison with the teams in the right part of the graph.

The axis F2 – administrative capital – is formed by the teams number thirteen and twelve.

On the one pole we see a team led by a man with a large number of administrative assets: he is a

member of academic boards of the Russian Foundation for Basic Research, of the the Ministry of

Education and Science of the Russian Federation, of the Higher Attestation Commission of the

Russian Federation; he is in the forefront of his own movement in the Russian Academy of Sciences

and has established a scientific school on basis of the laboratory he is a head of.

20

The team number twelve, with its little administrative capital, is on the other pole of this

axis. As a matter of principle, the teams that are located in the lower part of our graph, as opposed to

the ones in the upper part, have large administrative assets, a bigger number of employees that hold

positions in scientific, educational and state institutes, are members of editorial and university

boards, working groups and various committees.

As we differentiate the R&D teams by two extra factors – namely, expert capital volume and

will to be integrated into the international scientific society, we can thus divide our teams in four

types that coincide with respective quadrants on the Fig. 8.

Only two teams ended up in the first quadrant (upper right corner of the figure), the

number ten and the number nine. They are international science-oriented, geared for integration in

the international scientific society, have a high annual number of reviewed publications in foreign

magazines. Moreover, the team number nine has registered the highest number (25) of patents

among all the research groups. Both number ten and nine are working on implementation of large-

scale projects in fundamental research.

The second quadrant (lower right corner) gathered R&D teams that are characterized by

large volumes of scientific and expert capitals (number 11-formal, number thirtenn, number three

and number seven). The leaders of these teams have high citation indexes (and during the

interviews, they were actually able to name them) and actively participate in various conferences.

Moreover, they are often invited as experts and are members of Russian and international

programmes for science development. For example, researchers of the team number seven belong to

eight expert councils, three of which are international. The leader of the team number 11-formal is a

member of editorial boards of ten international scientific magazines.

In the third quadrant (lower left) we see four teams: number four, six, eight and five. They

are Russia-oriented: their articles have mainly been published in Russian magazines and

compilations, they participate in conferences and seminars in Russia, are not embedded in the

international scientific ccooperation system and international professional associations. Their

publication activity is rather low, they are mainly concerned with applied fields, and their projects

are mostly financed through contracts they signed with Russian organizations.

The fourth quadrant (upper left) gathered four teams (number 1, number two, number 12

and number 11-informal). Among those, the number 11-informal is rather an exception, and the only

common feature it shares with other teams in the same quadrant, is educational activity: all four of

these R&D teams are linked to the higher education system in Russia (namely, to the

21

LomonosovMoscow State University, Mendeleev University of Chemical Technology of Russia and

private higher educational institutions) and science pursued in the universities. The teams number

one, two and twelve are mainly geared for the Russian scientific achievements, ahev low publication

activity and are not integrated into the expert community. Unlike the teams in the third quadrant,

these are less concerned with applied science.

Efficient management of collective human capital

Achieving our goals, we had to elaborate a rather detailed quantitative analysis which leads

to construction of incredibly long logic inferences. This multitude, in turn, prevents us from being

able to choose rationally and consciously just one possible alternative, unless we recur to symbolic

means that would help us formalize those processes and create an algorithm for them. Mathematics

offers us a plenty of similar symbolic instruments, and as a consequence, our R&D team human

capital model is highly mathematized.

Since we conceptualized human capital of a research and development team as a distribution

function of social differences, we revealed a general principle that can be used as an instrument for a

fine and compact analysis of limitations for human capital management in specified conditions. In

mathematics this principle is called the principle of stationary action (or the principle of the least

action). It postulates that R&D team’s own human capital (that is, the one that corresponds to the

optimum level of efficient management), stands out of the row of all other possible human capitals

for the same team just given the condition that its action is stationary. In our case the principle of

stationary embodies the sociological idea of stable interaction and balance between the three ‘acting

reasons’:

Spontaneous development of R&D team’s human capital;

Purposive control actions;

The human capital is determined by the social conditions of scientific production.

Optimality of one’s own human capital is guaranteed by a certain balance between

management, control, social conditions and prerequisites, and spontaneous activities of researchers

[Kitova, 2014]. Naturally, optimality of any model of collective agent’s human capital management

in the scientific field is relative, as it’s optimal only for certain conditions and will cease being

optimal under other conditions. Whatever model we choose, there will always be a whole number of

circumstances that were not taken into consideration when it was constructed – and thus, its

optimality can be questioned. In other words, it can always be proved that an adopted criterion is a

22

criterion for an optimal solution only if certain assumptions and limitations worked out. The

problem of optimality of social control of a R&D team is not in the limitations of a mathematic

model. Instead, it hides in the scientific practices themselves. There are no management strategies

that would be unconditionally reliable in all situations, and there are no complete deadlocks. That’s

why we do not pretend that our model of optimal human capital for a scientific research and

development team can find some sort of absolute extremum. To the contrary, in the first place, our

conceptual model is an instrument for analysis of the current situation in the scientific production in

Russia.

Figure 9:Ranking under efficiency of management

It turned out that R&D groups’ human capital values that correspond to optimal management

form a sufficiently broad class of distributions that coincide with Pearson distribution. Based on

that, we defined the concept of efficient management of collective human capital as a probabilistic

distance from an empiric distribution function of social differences to a theoretical distribution

function that corresponds to optimal management.

23

Figure 10:Ranking under social potential

Of course, the sociological concept of collective human capital is supported by the fact it can

be applied to analysis of empirically detected social phenomena. After we changed our sample, we

proved that human capital management efficiency correlates with two indicators of R&D team

functioning: with the number of post-graduate dissertations (the first stage of a doctorate in Russia)

presented by researchers, post-graduate and PhD students of a team, and with the number of articles

published in cooperation with colleagues from other countries. Such result seems very logical and

proves the relevance of our definition: efficiency of human capital management cannot be

independent from the production of human resources in the scientific field. The significance of

international scientific cooperation can be easily proved too.

We created a space for the R&D teams we have examined using multidimensional scaling of

empiric functions. Its structure can be described through distribution of specific capitals that

characterize concordant changes of key (active) properties of members of all R&D teams. From the

empiric point of view, every such capital represents a statistically constructed factor that embodies

some invariant quality of research and development groups’ space that depends on the distribution

of power and authority in that space. Two axes of this scientific space correspond to volume of

aggregate scientific and administrative capitals possessed by a R&D team

Distribution of scientific and administrative capitals among the R&D teams in question

subordinate directly measured active properties, and thus for research purposes we can substitute a

description of a large number of ‘microscopic’ variables by a description of two or three substantial

‘macroscopic’ ones.

24

Our research shows that factors that structure the space of team leaders of the R&D teams

we studied do not coincide with the factors that determine the space of the R&D teams themselves.

Even though the contribution of heads of research groups to the distribution of active properties of

their teams is very big, as they have the biggest values of all measured sociological indicators, this

contribution is not crucial. The space of the R&D teams has a far more complicated structure. A

head of a team would typically seek recognition in the scientific world (which is revealed, amongst

other indicators, in a high citation index), while the space of R&D teams is structured by such

factors as cumbersome functions of active properties which we treat as aggregate volume of

scientific and administrative capital possessed by the research teams.

It is worth mentioning that the efficiency of human capital management proves significant

correlation between social potential of a R&D team that is our axis for the R&D teams space

constructed through multidimensional scaling of weighted average social differences between

members of research teams. The weighted average were calculated on basis of publication activity

and recognition in the scientific world. In other words, the more efficient social control is, the lower

is the social homogeneity and the sharper the unevenness of the active properties distribution is, and

respectively, social practices are further divided.

In such a way, efficiency of human capital management depends on average properties of

social composition of a collective agent in the scientific production.

Conclusion

With bigger or smaller success, common human capital management practices in Russia

make use of methods of centralized external action on a R&D team. However, Russian experience

proves that scientific achievements are relatively inefficiently regulated if exposed only to external

administrative and financial measures. Thus there appeared a need to switch the focus to target

influence that would be able to form R&D team’s own structures that spring up on their own,

besides external political or economic factors [Afanasyev et al., 2014]. These structures are invariant

with respect to various changes of research practices’ social conditions. It is clear that R&D team’s

own structures are ‘by default’ relevant to the essence of production of scientific knowledge and can

function as prerequisites for breakthrough innovations.

25

Therefore, we shall not only fully take into consideration immanent properties of a R&D

team, but, further on, to base our analysis on the following indisputable fact: a R&D team enjoys

such a property as directed self-organization. That means that structures and changes will arise in a

research group only when they are already potentially present there. Only when structures are

mature enough, control action and social conditions will allow research activities to reach the next,

completely new level. Respectively, the scientific policy does not aim to construct any pioneering

innovatory scientific organizations from the scratch. Instead, it should aim at activation of ‘hidden

opportunities’ of scientific organization, to enforce its own structures and development path. We can

even say that social control of a R&D team is then functioning rationally and reasonably when it is

based on natural course of development of a team and does nothing but introduces slight corrections

to this natural events instead of arbitrary unconditioned social construction based on some dogma

assimilated earlier.

Therefore, systematic record-keeping with respect to internal degrees of freedom of a R&D

team, spontaneity and self-organization abilities of the latter imply, first of all, significant correction

of goals set for social control over a R&D team. Second, we witness formation of new social and

economic feedback mechanisms. More precisely, we are saying that management and control of a

R&D should principally be aimed at reaching a “goal attractor”. The latter represents both multiple

conditions towards which a R&D team is heading asymptomatically in the result of its natural

evolution and a management regime desired by the control component. Apart from desired

economic, financial and other research procedures, the goal attractor should reflect the scientific

essence of a R&D team. The fact that R&D team’s own structures and goal attractors are included

explicitly in the control and management development procedures provides the team with an

opportunity to create a feedback system analytically, both for positive and negative feedback.

Acknowledgement: The author wishes to express her sincerest

gratefulness to National Research University Higher School of Economics

Basic Research Fund’ kind support for conduting this work.

26

References:

Afanasyev M., Meissner D., Cervantes M. (2014) Towards FET Concept: Pathway to Evaluation of

Foresight Effectiveness, Efficiency and Validity / Working papers by NRU Higher School of

Economics. Series WP BRP “Science, Technology and Innovation”.№ 31/STI/2014.

Barnes B. (1982) T.S. Kuhn and social science. – London: Macmillan.

Becker G. (1964) Human capital: theoretical and empirical analysis, with special reference to

education. N.Y.

Bellotti E. (2011) “The social processes of production and validation of knowledge in particle

physics: Preliminary theoretical and methodological observations”, in Procedia — Social

and Behavioral Sciences, vol. 10, № 0, pp. 148–159.

Blaug M. (1976) The empirical status of human capital theory: a slightly jaundiced survey, in

Journal of Economic Literature. Vol. 14.№3.

Bloor D. (1976) Knowledge and social imagery. – Chicago, Il.: University of Chicago Press.

Bloor D. (1983) Wittgenstein: a social theory of knowledge. – New York, N.Y.: Columbia

University Press.

Bordley R. F. (1983) “A central principle of science: Optimization”, in Behavioral Science, vol. 28,

№ 1, pp. 53–64.

Bourdieu P. (1975) “The specificity of the scientific field and the social conditions of the progress of

reason”, in Social Science Information, vol. 14, pp. 19–47, December.

Bourdieu P. (1984) Homo academicus. – Paris : Éditions de Minuit.

Bourdieu P. (1988) Homo Academicus.– Stanford, Cal.: Stanford University Press.

Bourdieu P. (1984) Distinction: A Social Critique of the Judgment of Taste. – Cambridge, Mass:

Harvard University Press.

BourdieuP. (1985) “The social space and the genesis of groups”, in Theory and Society, vol. 14,

pp. 723–744, November.

Bourdieu P. (1991) “The peculiar history of scientific reason”, in Sociological Forum, vol. 6, pp. 3–

26.

Bourdieu P. (1997) Les Usages Sociaux de la Science: Pour une Sociologie Clinique du Champ

Scientifique. Paris: Les Editions INRA.

Bourdieu P. (2001) Science de la science et réflexivité: Cours du Collège de France, 2000–2001. –

Paris: Éd. Raison d’agir. P. 17–19.

27

Bourdieu P. (2002) “The forms of capital,” in Readings in Economic Sociology (N. W. Biggart, ed.),

pp. 280–291, Malden, Mass: Blackwell.

Bourdieu P. (2004) Science of Science and Reflexivity.– Chicago, Il.: University of Chicago Press.

Bourdieu P. (1979) La Distinction. Critique sociale du jugement. – Paris: Les Éditions de Minuit.

Bozeman B., Corley E. (2004) “Scientists’ collaboration strategies: Implications for scientific and

technical human capital”, in Research Policy, vol. 33, no. 4, pp. 599–616.

Bozeman B., Dietz J.S. and Gaughan M. (2001) “Scientific and technical human capital: An

alternative model for research evaluation”, in International Journal of Technology

Management, vol. 22, № 7/8, pp. 716–740.

Brosnan C. (2011) “The significance of scientific capital in UK medical education,” in Minerva, vol.

49, pp. 317–332.

Brubaker R. (2005) “Rethinking classical theory”, in After Bourdieu: Influence, Critique,

Elaboration (D. Swartz and V. Zolberg, eds.), Dordrecht; Boston, Mass: Springer

Netherlands. Pp. 25–64.

Camic C. (2011) “Bourdieu’s cleft sociology of science”, in Minerva, vol. 49, pp. 275–293.

Chalmers A.F. (1990) Science and its fabrication. – Milton Keynes, U.K. : Open University Press,

1990.

Chernyak Y. B., Leonov A. I. and Lerner A. Y. (1992) “Toward the constructive theory of human

social behavior. I. General principles,” Annals of the New York Academy of Sciences, vol.

661, №1, pp. 256–276.

Collins H.M., PinchT. J. (1982) Frames of meaning: the social construction of extraordinary science.

– London: Routledge&Kegan Paul.

Coradini O. L. (2010) “The divergences between Bourdieu’s and Coleman’s notions of social capital

and their epistemological limits”, in Social Science Information, vol. 49, pp. 563–583,

December.

Corolleur C. D., Carrere M. and Mangematin V. (2004) “Turning scientific and technological

human capital into economic capital: The experience of biotech start-ups in France”, in

Research Policy, vol. 33, № 4, pp. 631–642.

Dietz J.S. and Bozeman B. (2005) “Academic careers, patents, and productivity: Industry experience

as scientific and technical human capital,” in Research Policy, vol. 34, № 3, pp. 349–367.

Gackstatter S.; Kotsemir M.; Meissner D. (2014): Building an Innovation-Driven Economy – The

Case of BRIC and GCC Countries. Foresight, Vol. 16, Iss. 4, pp. 293-308 (2014).

28

Garforth L. and Kerr A. (2011) “Interdisciplinarity and the social sciences: Capital, institutions and

autonomy”, in The British Journal of Sociology, vol. 62, №4, pp. 657–676.

Gokhberg L., Kuznetsova T., Roud V., Zaichenko S. (2013) Monitoring innovation activities of

innovation process participants (2011: R&D Organisations) Basic Research Program

Working Papers Series: Science, Technology And Innovation WP BRP 06/STI/2013.

Grossetti M. (1986) “Metaphore economique et economie des pratiques”, inRecherches

Sociologiques, vol. 17, no. 2, pp. 233–246.

Hagstrom W. O. (1975) The Scientific Community. – Carbondale, Il.: Southern Illinois University

Press.

Hong W. (2008) “Domination in a scientific field: Capital struggle in a Chinese isotope lab”,

inSocial Studies of Science, vol. 38, pp. 543–570, August.

Jain S. (2013) “Bourdieu’s sociology: A post-positivist science”, in Thesis Eleven, vol. 117, №1,

pp. 101–116.

Karady V. (1977) Titres, diplômes et carrières scientifiques universitaires dans la France du 19e

siècle. – Paris, Centre de sociologie de l’éducation et de la culture.

Karady V. (1979) “Strategies de la reussite et modes de faire-valoir de la sociologie chez les

durkheimiens”, in Revue française de sociologie. 1979, Vol. XX.№ 1. P. 49-82.

Kitova G. (2014) Taxation of R&D: review of practices Basic Research Program Working Papers.

Series: Science, Technology and Innovation. WP BRP 28/STI/2014.

Lebaron F (1997) “La dénégation du pouvoir”, in Actes de la recherche en sciences sociales, № 119.

Lebaron F. (2001) “Economists and the economic order: The field of economists and the field of

power in France,” in European Societies, vol. 3, № 1, pp. 91–110.

Lebaron F. (2009) “How Bourdieu “quantified” Bourdieu: The geometric modelling of data,” in

Quantifying Theory: Pierre Bourdieu (K. Robson and C. Sanders, eds.), Dordrecht: Springer

Netherlands. Pp. 11–29.

Lin M.-W. and BozemanB. (2006) “Researchers’ industry experience and productivity in university

industry research centers: A “scientific and technical human capital” explanation”, inThe

Journal of Technology Transfer, vol. 31, pp. 269–290.

Lundvall, B.-Å.(1985), Product Innovation and User–Producer Interaction.Aalborg University Press.

Lundvall, B.-Å. (1992), National Systems of Innovation: Towards a Theory of Innovation and

Interactive Learning, London: Pinter Publishers.

29

Lundvall, B.-Å., Johnson, B., Andersen, E.S., and B. Dalum (2002), “National systems of

production, innovation and competence building”, in Research Policy, Vol. 31, pp. 213–231.

McGuire W. L. (2011) Constructing Quality in Academic Science: How Basic Scientists Respond to

Canadian Market – Oriented Science Policy — A Bourdieusian Approach. PhD thesis, Dalla

Lana School of Public Health, University of Toronto, Toronto.

Meissner D., Roud V., Cervantes M.(2013) Innovation Policy or Policy for Innovation? – In Search

of the Optimal Solution for Policy Approach and Organisation, In: Meissner D., Gokhberg

L., Sokolov A. (eds.) (2013) Science, Technology and Innovation Policy for the Future -

Potentials and Limits of Foresight Studies. Springer, Heidelberg/ New York/ Dordrecht/

London.

Merton R.K. (1979) The Sociology of Science: Theoretical and Empirical Investigations. Chicago,

Il.: University of Chicago Press, 2nd ed.

Mincer J. (1974) Schooling, Experience and Earnings. – N.Y.

Mincer J. (1994) The Production of Human Capital and the Life cycle of Earnings: Variations on a

Theme // Working Paper of the NBER, No 4838 (Aug.).

OECD (2002) Frascati Manual: Proposed Standard Practice for Surveys on Research and

Experimental Development. Paris.

Panofsky A. (2011) “Field analysis and interdisciplinary science: Scientific capital exchange in

behavior genetics”, in Minerva, vol. 49, pp. 295–316.

Romer P. M. (1990) “Endogenous Technological Change”, in Journal of Political Economy,

№98(5), pp. 71–102.

Ruget V. (2002) “Scientific capital in American political science: Who possesses what, when and

how?”, in New Political Science, vol. 24, № 3, pp. 469–478.

Schultz T. (1971) Investment in human capital: the role of education and of research. N.Y.

Sismondo S. (2011) “Bourdieu’s rationalist science of science: Some promises and limitations”,

Cultural Sociology, vol. 5, pp. 83–97, March.

Shmatko N. (2011) “Scientific Capital as a Driver of Researchers’ Social Mobility”, in Foresight-

Russia, vol. 5.№ 3. P. 18–32.

30

Name*: Natalia Shmatko

Russia, Moscow, Myasnitskaya str. 20, 101000

National Research University Higher School of Economics

Institute for Statistical Studies and Economics of Knowledge,

Department for Human Capital Studies

Head of Department

E-mail: nshmatko@hse.ru

* Corresponding author

Any opinions or claims contained in this Working Paper do not necessarily

reflect the views of HSE.

© Shmatko, 2014